I fucking love this computer (MacBook 12"). Great keyboard (although I admit it takes 2 or 3 days to get really used to it), great trackpad, great screen, comes with macOS, completely silent and fanless, perfect portability.
Post technology you love that /g/ loathes.
Battery life?
>>62085041
Pretty good. Haven't done any scientific testing with it, but it lasts all day for me. Reviews say it's around 9 hours average use (wifi, web, some video, etc).
>>62085014
Inb4 $1000 for core m3
>>62085117
Your paying for the God tier screen, God tier trackpad, God tier portability and miniaturization.
>>62085137
>God tier
Yeah, no. I'm sure that most people agree that a even a fucking Chromebook offers better value over this thing, especially if you're just going to use it as a terminal anyway.
Hell, if I'm going to spend that much, I'd just spend the extra 200 bucks and get a MacBook pro. It gives you a whole .5 mm of key travel and much bigger track pad, all in a footprint that's barely larger.
Its funny that you feel the need to validate your purchase. Enjoy your throttlebook.
>>62085297
*A whole extra 0.5 mm of keytravel
>>62085297
Thanks, I will!
>>62085014
It's useless without CUDA.
>>62085014
My MBP running Kali at work. People here shit on them cause they are broke jobless children.
>>62085396
Macs are good as Linux machines, the only problem is that you look like a gay with it.
And it doesn't have CUDA, you can't learn nor use GPU computing with it whilst heterogeneous computing is a backbone of linear algebra operations in large scale for example for training deep neural networks.
>>62085575
To be fair no one trains nets on laptops
I fucking love this computer (ThinkPad T60p). Great keyboard (Just like a desktop one), great trackpad+trackpoint, great 4:3 IPS screen, comes with linux, completely servicable, perfect durability.
>>62085623
So Apple users have to develop GPU algorithms directly on servers.
>>62085677
Everyone just trains on their desktops or cloud.
>>62085685
Training isn't the same as development, debugging and testing a GPU solution.
>>62085735
Most frameworks I've used are agnostic to CPU/GPU development i.e. the same code should work with CPU as with GPU.
>>62085750
I've just pointed out the limitation of Macs. DNNs are just an application there's more you can do with large matrices or large amount of matrices on a GPU.
>>62085014
MBPs are better Linux machines than ThinkPads.
You lied to me /g/